US2415416A - Froth flotation of silica from iron ore with anionic collector - Google Patents

Description

Patented Feb. 11, 1947 FROTII FLOTATION F SILICA FROM IRON 1 ORE WITH ANIONIC COLLECTOR Julius Bruce Clemmer and Milton Friel Williams, Jr., Tuscaloosa, Ala., assignors to the United States of America, as represented by the Scoretary oi the Interior No Drawing. Application February 9, 1945,

Serial No. 576,997

6 Claims.

(Granted under the act of March 3, 1883, as amended April 30, 1928; 37.0 0. G. 757) tion of the siliceous gangue constituents yields an iron enriched product; more particularly it relates to a froth flotation process employing anionic collecting agents to effect flotation of activated siliceous gangue constituents from caustic alkaline iron ore pulps while retarding flotation of the iron oxides with vegetable tannin extracts or acidic polyhydroxyaromatic compounds derived therefrom by hydrolysis.

An object of this invention is to provide a froth flotation process for concentrating iron oxides in pulps containing them in the presence of siliceous gangue. A further object is to provide a flotation process for separating silica from iron oxides employing anion-active collecting agents. A still further object is to provide a flotation process for concentrating iron ores and products fromthe milling of iron ores containing both calcareous and siliceous gangue materials. Still other objects include the development of a flotation process which will have greater selectivity in separating siliceous materials from oxide iron ore pulps and thereby efiect greater operating economies. Other objects, purposes, and advantages of the invention will hereinafter more fully appear or will be understood from the detailed description of its practice.

In a co-pending application, assigned to the same assignee as the present application, Serial No. 473,162, filed January 22, 1943, by Julius Bruce Clemmer and Ballard H. Clemmons, is described a process for beneficiating iron ores by froth flotation wherein anionic collecting agents are employed to float activated siliceous gangue constituents from caustic alkaline iron ore pulps with a pH of at least 10, while retarding flotation of the iron oxides with metaphosphates or polyphosphates.

As a result of further research and experimentation, we have discovered an improved method of concentrating iron=ores by froth flota-- tion of activated siliceous gangue constituents employing anionic collecting agents, such as the higher fatty acids, resin acids, mixtures of fatty and resin acids, and soaps thereof, to float the activated siliceous materials from caustic alkaline ironore pulps with a pH of at least as high as l0,'while retarding flotation of the iron oxides with vegetable tannin extracts or related acidic polyhydroxy aromatic organic compounds, such as catechol, hydroquinone, pyrogallol, gallic acid, and tannic acid, derived from vegetable tannins by hydrolytic decomposition.

Tannin is a class name applied to a group of astringent products of vegetable origin employed for the tanning of animal hides to yield leather. Tannins occur widely distributed in nature in the bark, wood, leaves, roots, andv fruits of various trees and plants. They are found in commercial quantities, for example, in the bark and wood of the quebracho, oak, chestnut, pine, hemlock, acasia, mangrove, mimosa, and other trees; in the leaves of sumac and gambier; the roots of canaigre and palmetto; the fruits of myrobalans and Turkish oak; and in gall-nuts, the warty masses which grow on trees as a result of insects puncturing the bark.

The tannins are extracted from these and other plants and trees by comminution of the tannin-bearing material, leaching with water, and recovering the extracted solidsby evaporation. The tannins are unavoidably diluted by other water soluble substances present in the bark or wood, such as mineral salts, carbohydrates, etc., and the extracts vary in the nature and proportion of tannin and non-tannin materials. The extracts may also !be contaminated with suspended insoluble matter and appear turbid or muddy. The crude extracts may be treated by different methods before evaporation to yield a purified extract. Suspended solids may be removed by filter-pressing, centrifuging, or by settling and decanting. Another method of treatment is to add blood albumin and heat he extract to about 70 C., at which temperature the albumin coagulates and carries down with it the suspended matter and some of the deeply colored non-tannin bodies. Other methods of purification involve. treatment with chemical agents, such as sulfur dioxide and sodium bisulfite. The purified extracts are usually evaporated under vacuum to yield concentrated liquid extracts or substantially dehydrated residues for marketing.

Commercial liquid extracts may contain 50 to percent, or more of water, and the powdered or solid commercial extracts may contain from 5 to 15 percent water.

Although the exact chemical structure of the These tannins give a distinguishing blue-blackcolor with ferric chloride, and yield no precipitate with bromine water. This class includes the tannins of gall-nuts, sumac, oak wood, and chestnut wood.'

(2) The catechol tannins hydrolyze to yield protocatechuic acid, which is convertible by heat into catechol. These tannins give a, distinguishing green-black color with ferric chloride, and

Q ties of other alkaline reagents, such -assodium sulfide, sodium carbonate, and sodium silicate,

yield a precipitate with bromine water.' This class includes the tannins of oak bark, hemlock bark, and quebracho wood. Arbutin, a glucoside, yields a catechol isomer, hydroquinone, and belongs to this class.-

(3) The ellagi tannins yield ellagic acid on hydrolysis with dilute sulfuric acid. This class includes the tannins of divi-divi, myrobalans, and valonia.

This simple classification is arbitrary, however, as many of the natural tannins are quite complex and appear to be mixtures or combinations of the various classes.

We have found that various vegetable tannin extracts of the above designated classes are effective depressants for iron oxides during anionic flotation of the activated siliceous materials. The tannins which we have used and found satisfactory in the practice of our invention include the liquid and solid extracts of quebracho wood, chestnut wood and bark, hemlock wood and bark, oak wood and bark, acacia (Borneo cutch), sumac, mangrove, and myrobalans. The liquid and solid extracts were used interchangeably in our method of flotation, and, based on their respective tannin contents, were about equally effective. The non-tannins in the various extracts were inactive diluents and exerted no deleterious effect on flotation of the silica or retardation of the iron oxides,

In the practice of our invention, we may also employ as iron oxide depressants the acidic polyhydroxy aromatic compounds resulting from hydrolytic decomposition of natural vegetable tannin materials. These compounds, which may also be prepared by well knownmethods of organic synthesis, include catechol, hydroquinone, pyrogallol, gallic acid, and tannic (di-gallic) acid. While we prefer to employ the natural tannin extracts because of their lower cost, the acidic polyhydroxy aromatic compounds may be used if desired. In describing our invention, we shall for convenience use the term vegetable tannin extract to include the liquid and solid forms of natural tannin extracts of commerce and the acidic polyhydroxy aromatic compounds which may be derived therefrom or prepared by organic synthesis.

A caustic alkaline pulp with a pH at least as high as 10, and preferably in the range of 10 to 12, is preferable in our method of flotation to facilitate effective retardation of the iron oxides by the vegetable tannin extracts. We prefer to employ caustic alkalies, such as sodium or potassium hydroxides, or their equivalents, to establish the desired pH for flotation. Moderate quantimay be used in conjunction with the caustic alkalies if desired. A combination of hydrated lime and soda ash has been successfully employed in the flotation of many iron ores and served the dual purpose of activating the siliceous materials to anionic flotation and establishing the desired caustic alkalinity for flotation by formation of caustic soda within the pulp. The optimum pH for anionic flotation of the activated silica and retardation of the iron oxides varies slightly for different-ores but generally falls within the range 10 to 12, and the proper pH for any particular case is best determined by experimentation. Poor flotation of the silica and incomplete retardation of the iron oxides invariably results if the pH of the pulp is much less than 10; a pulp pH greater than 12 is not particularly objectionable, but the froth is inclined to be voluminous and flotation of the silica is sluggish. We therefore prefer in our flotation method to keep the pH i the range of 10 to 12. A

It is essential in the practice of this invention that the siliceous gangue constituents in the iron ore be activated and made floatable by the anionic collecting agents. The siliceous gangue materials in certain iron ores, such as the calcareous red ores of the Birmingham district, Alabama, are naturally activated to soap flotation, presumedly due to the presence of lime and/or magnesia salts in the ores. Siliceous iron ores, as exemplified by the weathered red ores of the Birmingham district and most siliceous ores of the Lake Superior district, seldom contain naturally activated siliceous materials, and the. successful operation of our process requires that the silica in such ores be activated for anionic flotation. Extended research on a varietyof siliceous iron ores demonstrated that our method of flotation lends itself readily to the use of silica activatingagents. Many alkaline-earth and heavy metal salts exhibit the property of activating quartz and vari-' ous silicate minerals to soap flotation if the proper quantity is employed at an optimum pH. We have found that conditioning of the caustic alkaline iron ore pulp with a moderate quantity of an activator selected from the group consisting of soluble calcium, magnesium, barium, strontium, and lead salts suffices to activate the siliceous constituents to anionic flotation by the higher fatty acids, resin acids, or their derived soaps without adversely affecting retardation of the iron oxides by the vegetable tannin extracts.

In the practice of our invention, we prefer to employ hydrated lime for activation of the silica. Grinding or blunging the iron ore with sufflcient hydrated lime to establish a pulp pH of about 11 suffices for complete activation of the siliceous gangue in most iron ores. The usual quantity of hydrated lime required to activate the silica in iron ores varies from 1 to 4 pounds per ton of ore, but some ores containing acidic salts may require 8 pounds or more of lime per ton. Our experiments on a variety of calcareous and siliceous iron ores have revealed that only part of the hydrated lime added to the flotation pulp is consumed in activating the siliceous constituents to anionic flotation; part of the added lime may react with salts in the pulp and is precipitated as carbonate, another part is adsorbed by the iron oxides, still another part remains in solution to satisfy the solubility requirements of the water, and finally any excess hydrated lime added to the pulp remains undissolved. The lime I consuming capacity differs for diflerent iron ores.

and the proper quantity of lime for activation of the silica is best determined by experimentation. Excessive quantities of dissolved lime in the conditioned pulp should be removed before flotation as they increase collector requirements.

Although we have achieved by the practice 01' this invention good flotation of silica from iron ore pulps containing 100 parts per. million of hydrated lime in solution, equivalent to about 0.8 pounds of dissolved hydrated lime per ton of solids in our tests, we prefer that the flotation pulp be substantially free of dissolved hydrated lime. The dissolved lime may be removed either by washing the conditioned pulp with fresh water, or by addition of sufficient soda ash to precipitate the lime as insoluble carbonate. Brief conditioning of the iron ore with suflicient hydrated lime to give a pulp with a pH of about 11 generally suffices for complete activation of the silica. Subsequent addition of soda ash to the pulp, the usual requirements for most ores being 1 to 2 pounds per ton of solids, precipitates the lime remaining in solution and gives a pulp of the desired caustic alkalinity for flotation. Additional caustic soda may be added to the pulp, however, to establish and maintain the desired pH during flotation. Should an excess of hydrated lime be inadvertently added to the pulp to activate the silica, as evidenced by a pH of about 12.3, the excess is best removed by settling and washing the pulp one or more times with fresh water to reduce the pH to 11, whereupon soda ash may be added to precipitate the remaining dissolved lime in the pulp before proceeding with flotation. If desired, however, the

practice of our invention to enable an improved separation to be obtained. The various addition agents which we have employed and found useful include the metaphosphates: the polyphosphates; sodium silicate; starches of different botanical origin, such as potato, wheat, corn, rice, arrow-root or tapioca solubilized by heat and/or chemical treatment to yield starch solutions or gels of the ruptured granules; hydrolyzed starch products, such as British gum or dextrins; and various crude and purified lignin sulphonates such asv may be recovered from paper mill sulflte liquors. These auxiliary addition agents are advantageous on some iron ores in that they enable more effective retardation of the iron oxides with minimal quantities of the vegetable tannin extracts during anionic flotation of the activated siliceous materials.

The proportions of the various reagents employed in the practice of this invention are subject to considerable variation, and the optimum quantities are best determined by experimentation for any particular case. Purity of the separated products is a reliable guide for reagent adjustment. An excess of collecting agent promotes flotation of a portion of the iron oxides with the siliceous gangue, whereas a deficiency of collector results in incomplete flotation of the siliceous gangue. Conversely, an excess of vegetable tannin extract may result in retardation of a portion of the siliceous materials, whereas a deficiency permits flotation of some of the iron oxides with the silica. Control of the collector and tannin extract is not critical, however, and reasonable variation in the quantities employed is permissible without adversely afiecting the separation.

not obligatory in the practice of this invention.

Some lime or magnesia salts in the flotation pulp are permissible, and, in fact, are advantageous siliceous gangue materials from iron oxides include the higher fatty acids such as oleic acid or red oil, the purified or crude sodium oleates, fish oil fatty acids, fish oil soaps, and various crude and purified tallols and sulfate soaps de-- rived from sulfate paper mill black liquors. The tallols, which as mixtures of fatty and resin acids, are relatively inexpensive and are the preferred collectors in the practice of our invention.

While we have achieved satisfactory flotation of activated silica from iron ore pulps containing substantial quantities of slime by the practice of our invention, we prefer that the pulps be deslimed. Desliming of the iron ore pulp should be practiced whenever permissible as subsequent flotation of the activated silica is more rapid and complete, and less reagents are required.

We have discovered that various addition agents may be advantageously employed to supplement the vegetable tannin extracts'in the To avoid possible confusion in describing the results of our flotation tests, we shall hereinafter designate the iron enriched pulps from the roughing and cleaning steps as "iron concentrates, rougher" and "iron concentrates, cleaner, re-' spectively. The floated silica products from the roughing and cleaning steps will be designated as rougher froth and silica rejects, respectively. In describing the results of our tests we shall consider and report the iron enriched pulps from the cleaning steps (middlings) as finished concentrates. It will be readily apparent to those skilled in the art that the middlings may be retreated by conventional methods in continuous flotation op erations to yield an even higher grade concentrate without departing from the spirit of the invention.

Our invention will be further illustrated, but is not intended to be limited by the following examples of practice:

Example I A sample of calcareous red iron ore was obtained from an operating mine in the Birmingham district, Alabama. The ore was typical of the disto this invention, the iron ore or product to be treated is first ground to proper size for flotation,

if not already of such size, by conventional methdlings) the rougher iron concentrates (tailings), and the untreated slime were dried, weighed, and assayed. The results of the test follow:

Assay, percent Distribution, percent Product gs g Fe 0210 Insol. Fe 09.0 111501.

Iron concentrates:

Rougher 36. 4 52. 4 5. 1 11. 6 52. 5 17. 17. l Cleaner 11.8 33. 1 l5. 8 1'7. 2 10. 8 17. 2 8. 3 Slime 27. 0 43. 2 l0. 7 15. l 32. l 26. 6 16.

Composite 75. 2 46. 1 8. 8 l3. 7 95. 4 60. 8 41. 9 Silica rejects 24. 8 6. 9 17. 2 57. 7 4. 6 39. 2 58.1

Composite feed 100. 0 36. 3 10.9 24. 6 100.0 100. 0 100. 0

ods. The fineness of grind may vary from 35 to 200 mesh or finer, depending on the degree of liberation of the particular ore; substantially complete liberation of the iron oxides and siliceous gangue is essential for a satisfactory separation by flotation, and the fineness of grind should be selected accordingly. A limiting size of 100 mesh was chosen for this ore.

A 250-gram portion of the ore rolls-crushed to 20 mesh was wet ground to pass 100 mesh using 25 pounds of one-half inch steel rods as the grinding media with 250 m1. of tap water in a laboratory rod mill. The ground charge was diluted to a volume of 2.5 liters with additional tap water to give a pulp containing about 10 percent solids. The pulp which was flocculated was then dispersed using caustic soda and soda ash equivalent to 2.0 and 1.0 pounds per ton of ore, respectively. The dispersed charge was fractionated by sedimentation and decantation to remove the bulk of the slime finer than microns. The granular portion, essentially free of slime and coarser than 20 microns, was then repulped with additional tap water and transferred to a small mechanical flotation cell of standard design. Sufiicient tap water was added to give a pulp for flotation containing about 20 percent solids.

Flotation of the naturally activated silica was effected from a caustic alkaline pulp using tallol as the collector, and Purex SM brand, a quebracho extract marketed by The Tannin Corp., as the iron oxide depressant. The reagent charge expressed in conventional pounds per ton of ore was as follows:

The substantially deslimed pulp was first'conditioned with the caustic soda (commercial lye) and quehracho extract to establish the alkalinity desired for flotation and to retard the iron oxides. Tallol was chosen for the collector in the test. The pulp was conditioned withthe tallol, and air was then allowed to enter the cell and resulted in immediate formation of a compact, heavily mineralized froth of the siliceous materials. The froth was collected for 1.5 minutes, whereupon flotation was complete. The rougher froth was cleaned twice by re-fioating in the same cell using tap water with additional caustic to maintain alkalinity. The final silica rejects, the combined iron concentrates from the cleaning steps (mid- The flotation silica rejects accounted for 24.8 weight percent of the feed, and contained 58.1

percent of the insoluble and only 4.6 percent of l the iron. The untreated slime and cleaner iron concentrates were sufliciently low in insoluble to be considered finished grade. The composited iron concentrates represented a recovery of 95.4 percent of the iron in the feed and assayed 46.1 percentFe, 8.8 percent CaO, and 13.7 percent insoluble.

The results of the recorded test are typical of those obtained on deslimed charges of the calcareous red ore ground to pass 65, 100, or 200 mesh in a pebble mill, iron ball mill, or iron rod mill. Good flotation of the activated silica was achieved by using a moderate quantity of tallol, sulfate soap, oleic acid, sodium oleate, or similar anionic collecting agents in caustic alkaline pulps in conjunction with the Purex SM Brand quebracho extract to retard the iron oxides. A moderate quantity of the quebracho extract suiiiced for the separation and control of the depressant was not critical.

Example II The summarized results of several flotation tests on the calcareous red ore using various vegetable tannin extracts as the iron oxide depressant are given below. The ore charges were ground to pass mesh, deslimed at 20 microns, and floated by the procedure described in Example I. The tests Were made employing 1.0

pound per ton of the designated vegetable tannin extract together with 0.8 pound of tallol per ton as the collector, and 2.0 to 3.0 pounds per ton of caustic soda as needed to give a pulp pH of about 11 for the roughing operation. The rougher froths were double-cleaned in the tests using additional caustic soda and tannin to retard the remaining iron oxides; the quantities employed in the cleaning steps were 0.8 and 0.1 pound per ton, respectively. The grade of the composited flotation iron concentrates from the roughing and cleaning steps and the percentage of iron recovered from the flotation feed in the tests are as follows:

, Assay, per cent Re Test Vegetable tannin Weight, covcry, No. extract employed per cent per cent Fe CaO Insol. Fe

Flotation feed 100.0 33. 4 11. 4 28. 2 100. 0 l C, F, del C" brand 78. 2 41.6 11. 8 l1. 6 97. 4

quehracho extract. 2 Hy-Test" brand que- 66. l 47. 7 7. 7 13.0 93. 1

bracho extract. 3"--. Oak bark extract 72. 7 44. 2 10. 5 12. 5 96.1 4- Borneo cutch 79. 9 41. 5 ll. 7 l4. 8 97.9 5 Hemlock extract- (i6. 5 47. 4 9.3 12.1 93. 9 6 Chestnut extract. .i 69. 9 45. 4 10.2 12. 1 94. 8 7. Myrobalan'extract 65.1 47. 4 8. 4 12.8 93.5

The various tannin extracts employed in the 9 tests were oi the dry. powdered forms and were selected as being typical of the galiol. catechoi, and ellagi classes of tannins commercially available. Based on their respective tannin content,

relatively low cost, the acidic poiyhydroxy aromatic compounds resulting from hydrolysis of natural or treated tannin extracts of commerce, unprepared by organic synthesis, may likewise be to maintain the pulp alkalinity in each step. The grade of theflotation concentrates from the roughing and cleaning steps and the percentage of iron recovered from the flotation feed in the various tests are as follows:

Recovery, per cent Inscl. Fe

Polyhydroxy com- Assay per cent pound employed as depressant Weight, per cent Flotation feed Gallic acid Tannic acid (di-gallic acid) Catechol; Hydroquinone Pyrogallol the various commercial extracts were equally er- 5 employed without departing from the spirit of iective for retarding the iron oxides during our invention. anionic flotation oi the silica. Example IV 1 Example III A 250-gram charge of the calcareous iron ore was ground to pass 100 mesh as in Example I. w meme ferred to a small mechanical flotation cell of extracts as iron oxide depressants in our method of silica flotation. We shall now consider the :5 g fi g i was results of similar flotation tests on deslimed porg 22 8 ve i 1 mung tions of the calcareous red ore wherein various g percent so otat the silica polyhydmxy benzenes and benz 01 c acids were mm the iron oxides was eflected by the following employed to retard the iron oxides. These polyreagelgs eipressfed conventional terms or hydroxy aromatic compounds are closely related poun per on o to the vegetable tannins and may be derived from j y the extracts by hydrolysis of the glucosides. Acid hydrolysis or fusion of the dry extract with alkali yields the hydroxy acids which may be recovered. 1 2 1 2 3 The acids decompose on heating to yield the polyhydroxy benzenes. The compounds may also be g conveniently prepared from benzene and its de- 05 110 rivatives by well known methods of organic i gfg 3 2 M 2'5 M M synthesis. Pulp pH 11.0 11.0 10.8 10.85 10.8 10.8-

The flotation charges were ground to pass 100 mesh, deslimed at 20 microns, and floated by the The Slime-bearing P was conditioned with procedure given in Example L The comparative caustic soda (commercial lye) and quebracho 8X- tests were made using 1.0 pound per ton of the tract Tallol w then added and the P p a ain designated polyhydroxy compound to retard the bri fly conditioned as indicated. The rougher iron oxides together with 0.8 pound per ton of froth w triple-cleaned y -flo in the tallol as the silica collector and 2.0 to 3.0 pounds e 08 1 ng tap water for dilution together per ton of caustic soda as needed to establish a it Caustic o The rougher a d OOmP tEd .pulp pH of about 11 for the roughing separation. cleaner iron concentrates and final silica rejects The rougher froths in the tests were triplewere dried, weighed, and analyzed. The results cleaned using 0.8 pound per ton of caustic soda of the test are as follows:

Product w g Assay,percent Distribution, percent percent Fe CaQ Insol. Fe CaO Insol.

Iron concentrates:

I Rougher 62.3 49.4 3.8 18.3 70.7 18.5 39.3 Cleaner 25.2 34.1 10.1 15.1 26.3 43.8 18.1

Composite- 80.5 44.0 as 11.4 91.0 02.3 67.4 Silica rejects 19-5 5.6 20.8 53.1 3.0 37.7 42.6

Composite ieed 100.0 30.5 10.8 24.3 100.0 100.0 100.0

Flotation of the silica from the slime-bearingv pulp rejected 42.6 percent of the silica (insoL) with a loss of only 3.0 percent of the iron. It is therefore apparent that our method of flotation is applicable to iron ore pulp containing substantial quantities of slime.

Example V In the previously described examples of practice, various vegetable tannin extracts and acidic polyhydroxy aromatic-compounds which may be derived from the tannin extracts by hydrolysis were employed to retard the iron oxides during anionic flotation of the naturally activated silica from either deslimed or total charges of the ground calcareous red ore. We shall now consider the results of flotation tests wherein auxiliary addition agents are employed to supplement the vegetable tannin extracts in our method of flotation.

- A 250-gram charge of the calcareous red ore ground to pass mesh in a laboratory rod mill and the charge including slime was floated by the procedure described in Example IV. Potato starch solubilized with caustic soda to rupture the granules at room temperature and form a ll starch solution was used as an auxiliary reagent in the test to supplement the quebracho extract and aid retardation of the iron oxides. The quantities of reagents employed in the test,

agents which have been employed in conjunction with the vegetable tannin extracts in the practice of this invention include the metaphosphates, the polyphosphates, and various crude expressed in conventional Pounds per o of 5 and purified lignin sulphonates derived from pawere as follows: per mill sulfite liquors. We shall now consider the results of a typical test using Goulac, a com- Condmmm Cleaner mercial calcium lignin sulphonate product from Reagent Rough 1O dehydration of paper mill sulflte liquor marketed N04 N01 N04 by the American Gum Products Company, in

conjunction with Purex SM brand quebracho exgarx esggfidgrgfiaafia. tract to retard the iron oxides during flotation braclio extract of activated silica from a calcareous red iron ore. The ore was ground to pass 100 mesh and the Timamim 2- charge, including slime, was floated by the pro- Pulp PH 10'75 cedure described in Eample IV. The quantities The results of the test are as follows:

of caustic, Purex SM brand quebracho, Goulac, and tallol employed in the roughing step were w i M Assay, percent Distribution, percent 8 8 Product per new;

Fe CeO Insol. Fe CaO Insol.

lron concentrates:

Rougher 52. 3 50. 4 5. 6 14. 4 73. 0 25. 5 30. 5 Cleaner 27. 8 31. 5 21. 6 l2. 6 24. 2 52. 7 14. 0

Composite 80. 1 43. 8 11. l 13. 8 97. 2 78. 2 44. 3 Silica rejects 19. 9 5. 1 l2. 5 69. 8 2. 8 21. 8 55. 7

Composite feed 100. 0 36. 2 ll. 4 25. 0 100. 0 100. 0 100. 0

The combination of quebracho extract and potato starch facilitated efiective retardation of both the slime and granular iron oxides during anionic flotation of the silica. Additional quebracho supplemented by starch was employed in the cleaning steps to facilitate recovery of calcite in the composited iron concentrates and.

yield a substantially self-fluxing iron product. If desired, however, the calcite may be floated with the silica and rejected by use of a small additional quantity of collector in the cleaning steps together with reduced quantities of tannin extract.

Other cereal and tuber starches, such as corn, wheat, rice, arrow-root, and tapioca, solubilized by heat and/or chemical treatment to yield solutions or gels of the ruptured starch granules were employed in other tests on the ore to supplement qllebracho and other vegetable tannin extracts. The results of the tests were similar to those recorded when optimum quantities of the starch and tannin were employed. The utility of the starches and hydrolyzed starch products for retarding iron oxides during anionic flotation of activated silica from caustic alkaline iron ore pulps is described in a co-pending application, Serial No. 567,763, filed December 11, 1944, by Julius Bruce Clemmer and Milton Friel Williams, Jr., and assigned to the same assignee as the present application. Although the starches and hydrolyzed starch products are not obligatory in the practice of this invention, they are advantageous addition agents and enable an im proved retardation of the iron oxides with reduced quantities of the vegetable tannin extracts.

Example VI In addition to the starches and hydrolyzed starch products, other useful auxiliary addition 3.0, 1.0, 2.0, and 0.8 pounds per ton of ore, respectively. The rougher froth was triple-cleaned using 0.8 and 0.04 pound per ton of caustic soda and quebracho, respectively, in each step. The composited rougher and cleaner flotation concentrates from the test assayed 46.6 percent Fe, 8.7 percent CaO, and 12.7 percent insol., and represented a recovery of 95.0 percent of the iron in the ore. The'silica rejects assayed 6.5 percent Fe, 18.2 percent CaO, and 57.2 percent insol., and contained 63.2 percent of the silica, (insoluble) in the ore.

Example VII 5 Conditioner Cleaner Reagent Bougher No. 1 No. 2 No. i No. 2

Caustic soda l. 25 0.5 0. 5 Purex SM brand que- 10 bracho extract 1.00

Sodium hexametaphosphate 0.16 0 04 0.08

Sodium oleate. 2. 00 Oleic acid- 0. 96 Time, min 2. 5 2. 5 2. 5 2. 5 2. 5 Pulp pH 11. 1 11.1 10. 05 10. 8 10. 9

The results of the test are as follows: 7 and give the final silica rejects. The quantities w 1 M Assay, percent Distribution, percent es Product per cent Fe CaO Insol. Fe CaO Insol.

Iron concentrates:

' Rougher 78.8 48.2 8.5 12.2 89.3 74.1 49.7 Cleaner 10. 7 a0. 17. 0 1a. 0 0. 0 20. 1 7. 2

Composite s9. 5 4e. 7 0. a 12. 3 0s. a 04. 2 5e. 0 Silica rejects 10. 5 e. 9 a0 79. 4 1. 7 5.8 43.1

Composite feed 100.0 42.6 0.0 10.3 100.0 100.0 100.0

Example VIII The flotation tests heretofore reported were made on typical calcareous iron ores and mill products which contained siliceous gangue materials naturally activated to anionic flotation due to the presence of soluble lime or magnesia salts in the ores. We shall now consider the application of our invention to iron ores containing unactivated siliceous gangue constituents, The weathered and siliceous red ores of the Birmingham district, for example, seldom contain suflicient lime or magnesia salts to activate the contained siliceous materials to anionic flotation. In applying our invention to such ores the silica may be activated and rendered floatable by the use of hydrated lime or metal-salt activating agents.-

A sample of washer rejects was obtained from an iron'ore concentrator operating in the Lake Superior district, Minnesota. The sample as received was substantially finer than 48 mesh and represented a composite of classifier and dewater. ing device overflows impractical to treat by methods heretofore available. The sample contained hematite as the predominate iron oxide together with some specular hematite and magnetite. The gangue was mainly quartz with some iron silicates and clayey material. A head analysis of the sample gave 40.9 percent Fe and 33.8 percent insol.

A 250-gram portion of the washer rejects was wet ground to pass 100 mesh and deslimed at microns by sedimentation and decantation by the procedure described in Example I. The granular portion was transferred to a laboratory mechanical flotation cell and diluted with tap water to a pulp consistency of about 20 percent solids. The pulp was conditioned with hydrated lime to establish a pulp pH of about 11 and activate the silica to anionic flotation. Soda ash was then added to the pulp to precipitate the hydrated lime remaining in solution-and simultaneously form caustic soda in the pulp. The resulting pulp, substantially free of dissolved lime, was then conditioned with quebracho extract and tallol and subjected to flotation. The lime-activated silica floated readily from the retarded iron oxides in a compact, heavily mineralized, siliceous froth. The rougher froth was cleaned once by re-floating in the same cell to retard the remaining iron oxides of reagents, expressed in pounds per ton of ore, employed in the test were as follows:

Conditioner Hydrated lime Caustic soda Purex SM brand The grade of the flotation test products and the untreated slime fraction are as follows:

Distribution P Weight, Assay, per cent per cent roduct per cent Fe Insol. Fe Insoi.

Iron concentrates:

Rougher 28. 7 55. 8 9. 5 39. 5 8. l Cleaner 24. 3 51. 7 16. 1 31. 0 ll. 7

Composite. 53. 0 53. 0 i2. 5 70. 5 10. s Silica rejects 20. 5 l1. 9 80. 8 6. 0 49. 4 Slime (untreated). 26. 5 35. 9 39. 0 23. 5 30. 8

Composite- 100. 0 40. 5 33. 5 100. 0 100. 0

jected in the silica froth and 92.2 percent of the iron was recovered in the composited iron concentrates.

Example IX Another 250-gram portion of the siliceous washer rejects was ground to pass 100 mesh in a laboratory rod mill. The ground charge, including slime, was transferred to a small mechanical flotation cell, diluted with tap water to give a pulp containing about 22 percent solids, and subjected to froth flotation using the following reagent charge expressed in pounds per ton Reagent Hydrated lime Soda of feed:

Conditioner Cleaner Rougher No. 1 No. 2 No. 3 No. 4 No. 1 No. 2 No. 3

Time, mire Pulp pH 15 The grade of distribution of iron and insoluble in the test products are as follows:

Activation and anionic flotation of the silica of the silica (insoluble) in the washer rejects and yielded an iron enriched product which assayed 52.6 percent Fe and 14.5 percent insoluble and represented a recovery of 93.9 percent of the total iron. This test further demonstrates the utility of our method for floating silica from iron ore pulps containing substantial quantities of slime. While we prefer to employ deslimed pulps in the practice of our invention, desliming is not obligatory on many iron ores.

The results of the above recorded test were about average of those obtained on total charges of the ore. Various of the commercial tannin extracts were employed and found satisfactory for retarding the iron oxides during anionic flotation of the silica with oleic acid, sodium oleate, sulfate soap, tallol, and the fish oil fatty acids and soaps. Based on their respective tannin content, the liquid and solid tannin extracts, such as sulfited and non-sulfited quebracho extracts, oak extract, hemlock extract, chestnut extract, myrobalans, and Borneo cutch were about equally eflective for retarding the iron oxides. Control of the tannin and collector was not critical and reasonable variation in the quantities employed did not adversely aflect the separation, Considerable latitude was apparent both in the choice.

and quantities of tannin extract and collector employed.

Example X The utility of auxiliary addition agents, such as the solubilized cereal and tuber starches, metaphosphates, and lignin sulphonates, to supplement the vegetable tannin extracts and aid retardation of iron oxides during anionic flotation of naturally activated silica from calcareous iron ore pulps has been described in previous examples. These addition agents are likewise useful and give an improved separation on many siliceous iron ores. We shall now consider the. results obtained using causticized potato starch as an addition agent to supplement quebracho extract in the flotation of a siliceous ore.

A sample of tailing pond material from an operating plant in the Mesabi Range, Minnesota, was obtained for testing. The sample as received was substantially finer than 20 mesh and contained hematite as the predominant iron oxide together with some specular hematite and magnetite. The gangue was quartz together with some iron silicates and considerable clayey material. A head analysis of the sample gave 43.5 percent Fe and 31.8 percent acid insoluble.

A 250-gram portion of the sample was ground to 100 mesh and deslimed by the procedure given in Example VIII to reject the clay. The granular portion was transferred to a laboratory flotation from the slime-bearing pulp rejected 70.1 percent cell and diluted with tap water to give a pulp containing about 20 percent solids for flotation. The pulp was conditioned with hydrated lime to activate the silica and then with soda ash to precipitate the lime remaining in solution as carbonate. The lime-activated silica was floated using tallol as the collector while retarding the iron oxides with a combination or quebracho extract and potato starch. The potato starch was solubilized by rupturing the granules with caustic soda at room temperature to give a. solution containing 2.5 percent starch and 0.5 percent caustic soda. The reagent charge, expressed in pounds per ton, employed in the test was as follows:

Conditioner Reagent Rougher No. i No. 2 No. 3 No. 4

Hydrated lime 6. 0 Soda ash l. 5 Caustic soda-.. l. 0

Purex SM brand quebracho extract Potato starch Talloi Time, min Pulp pH The combination of tannin and starch was particularly efllcacious for retarding the iron oxides. A roughing treatment yielded a froth of sufliciently low iron content for rejection and a cleaning operation was unnecessary, The grade and distribution of iron and insoluble in the flotation products are as follows:

Example XI A sample of siliceous iron ore from a Michigan deposit was next examined. The ore resembled the weathered taconites of the Lake Superior district and consisted of hematite together with some magnetite finely disseminated in a siliceous gangue of quartz and various iron silicates. Grinding to mesh was necessary for liberation of the iron oxides.

A 250-gram portion of the ore rolls-crushed to pass 20 mesh was wet ground in a laboratory rod mill to 100 mesh. The ground charge including slime and grind water was transferred to the small mechanical flotation cell and diluted with tap water to give a pulp containing about 20 percent solids. The pulp was conditioned for 10 minutes with hydrated lime equivalent to 16 pounds per ton of solids to activate the quartz and iron silicates to anionic flotation. The pH of the lime-conditioned pulp was 12.3. The pulp was withdrawn from the cell and allowed to settle whereupon the clear supernatant water was carefully decanted to avoid loss of slime. The settled ore was repulped with fresh water to about 10 percent solids and allowed to settle. The lime-bearing water was decanted and the washing operation again repeated to give a final pulp (pH-9.8) substantially free of dissolved hydrated lime. The lime-conditioned and washed solids were returned to the cell and diluted with tap water to give a pulp containing about 22 17 percent solids for flotation. Flotation of the activated silica from the retarded iron oxides was achieved with the following reagentsyexpressed in pounds per ton of ore:

Conditioner Cleaner Reagent Roughcr No.1 No.2 No.1 No.2 No.3

Caustic soda 2. 0. 8 0. 8 0.8 Purex SM brand quebracho extract 0.08 0. 04 0. 04 31101.. 0.2 0.2 0.2 2. 5 2. 5 2. 5 l0. 7 l0. 8 l0. 8

The results of the test are as follows:

Distribution, weight, Assay, per cent per cent Product per cent Fe Insol. Fe Insol.

Iron concentrates:

Rougher 27. 7 57. 8 12. 0 41. l 8. Cleaner 33. 5 54. 5 16. 8 46. 9 14.

Composite 61. 2 56. 0 l4. 6 $8. 0 22. Silica rejec 38. 8 1 l2. 0 80. 0 l2. 0 77.

Composite ieed 100. 0 38. 9 40. 0 100. 0 100.

The results of the recorded test are typical of those obtained on lime-activated charges of the finely ground ore using various commercial tannin extracts as the depressants. Equally good actlvatiomof the quartz and iron silicate gangue minerals was achieved by either conditioning the pulp with a large quantity of hydrated lime and removing the excess dissolved lime by washing the pulp before flotation, or by conditioning the pulp with a lesser quantity of hydrated lime and precipitating the dissolved lime remaining in the pulp with soda ash prior to flotation as described in Example IX. Good flotation of the silica was achieved with moderate quantities of collector from both total and deslimed pulps of the ore containing not more than 100 parts per million of dissolved lime. Pulp'containing larger quantities of dissolved lime or calcium salts invariably required excessive amounts of collector and flotation of the silica was incomplete.

Example XII stituents to anionic flotation without impairing retardation of the iron oxides by the vegetable tannin extracts. Various ores require different quantities of the metal salt activating agent and I I 18 marlzed results showing the grade of the flotation concentrates from the tests are as follows:

The results of the previously described flotation tests in the various examples of practice are typical of those we obtained on a variety of calcareous and siliceous iron ores using various vegetable tannin extracts, or acidic Polyhydroxy aromatic compounds which may be derived from tannin extracts by hydrolytic decomposition, as the iron oxide depressants while floating activated siliceous gangue constituents from caustic alkaline iron ore pulps with oleic acid, sodium oleate, tallol, sulfate soap, or similar anionic collecting agents. While we have disclosed the pre ferred embodiments of our invention, it will be readily apparent to those skilled in the art that many variations and modifications may be made therein without departing from the spirit of the invention.

What is claimed:

1. A process for beneficiating iron ores containing siliceous gangue materials which comprises conditioning an aqueous pulp of the comminuted ore in the presence of hydrated lime to yield a Pulp with a pH of at least 11, thereafter washing the conditioned pulp with fresh water to yield a pulp containing not more than parts per million of dissolved hydrated lime, adding thereto a quantity of caustic alkali to establish a pulp pH at least as alkaline as pH 10, and a vegetable tannin extract, and then subjecting said pulp to agitation and aeration in the presence of an anion active collecting agent selected from the class consisting of higher fatty acids,

2. A process for beneficiating iron ores containing siliceous gangue materials which comprises conditioning an aqueous pulp of the comminuted ore in the presence of hydrated lime to yield a pulp with a pH of at least 11, adding thereto a quantity of soda ash to precipitate the dissolved hydrated lime and yield a pulp containing not more than 100 parts per million of hydrated lime remaining in solution, thereafter adding a vegetable tannin extract and an anion active collecting agent selected from the class consisting of higher fatty acids, resin acids, mixtures of fatty and resin acids, and soaps thereof, and subjecting said pulp to agitation and aeration whereby siliceous gangue is floated and beneficiated iron ore is depressed and recovered.

3. A process for beneficiating iron ores containing siliceous gangue materials which comprises addition to an aqueous pulp of the comminuted ore a quantity of caustic alkali to establish a pulp pH at least as alkaline as pH 10, and a soluble compound of an inorganic anion combined with a polyvalent metal selected from the class consisting of calcium, magnesium, strontium, barium, and lead, together with a vegetable tannin extract, and then subjecting said pulp to agitation and aeration in the presence of an anion active 19 collecting agent selected from the class consistini; oi higher fatty acids, resin acids, mixtures of fatty and resin acids, and soaps thereof, whereby siliceous gangue is floated and beneficiated'iron ore is depressed and recovered.

4. A process for the production of a beneficiated iron ore which comprises first subjecting-an aqueous pulp of such an ore to a preliminary activating treatment with a soluble compound of a polyvalent metal selected from the class consisting of calcium, magnesium, strontium, barium, and lead: and thereafter subjecting the thusactivated pulp to agitation and aeration at a caustic alkaline pulp pH at least as high as pH in the presence of a tannin substance selected from the group consisting of vegetable tannin extracts and polyhydroxyaromatic compounds de-- rivable therefrom by hydrolytic decomposition, and in the concurrent presence of an anion-active collecting agent selected from the class consisting of higher fatty acids, resin acids and their soaps, whereby siliceous gangue is floated and beneficiated iron ore is depressed and recovered.

5. A process for the production of a beneficiated iron ore whichcomprlses first subjecting an aqueous pulp of such an ore to a preliminary activating treatment with suiiicient lime to establish a pulp pH of at least 10, then incorporating with said pulp suillcient soda ash to precipitate excess lime while maintaining the desired caustic alkaline pulp for flotation, and thereafter subjecting the thus activated pulp of such an ore to agitation and aeration at a caustic alkaline pulp pH at least as high as pH 10 in the presence of a tannin substance selected from the group consisting of vegetable tannin extracts and polyhydroxyaromatic compounds derivable therefrom 20 by hydrolytic decomposition, and in the concurrent presence of an anion-active collecting agent selected from the class consisting of higher fatty acids, resin acids and their soaps, whereby siliceous gangue is floated and beneficiated iron ore is depressed and recovered.

6. A process for the production of a beneficiated iron ore which comprises first subjecting an aqueous pulp of such an ore to a preliminary activating treatment with excess lime, then removing the lime from the thus activated pulp to yield an activated pulp containing not more than 100 parts per million of residual dissolved REFERENCES crran The following references are of record in the file of this patent:

UNITED STATES PATENTS Number Name Date 2,364,618 Brown Dec. 12, 1944 2,341,046 Kirby Feb. 8, 1944